The study of human epileptic tissue has defined patterns of cell loss and reorganization in temporal lobe epilepsy. Clinical neurophysiological assessment of these patients prior to tissue removal has also defined patterns of spontaneous seizure morphology and propagation, but their basis is unknown. Basic research in animal models of epilepsy, taken together with these electrophysiological and histopathological findings, allows the development of a testable hypothesis to explain the peculiar patterns manifested by epileptic tissue. We propose that CA1 and dentate hilar cell loss provoked by febrile seizures create aberrant self- perpetuating excitatory limbic loops; that extrahippocampal lesions causing temporal lobe epilepsy are not associated with similar anatomical derangements; and that clinical neurophysiological measurements can predict the existence and location of such inherent hippocampal malfunction. Our methods to test this hypothesis, for which preliminary data provide considerable support, involve the correlation of the following discrete neurophysiological parameters with histopathology: 1) response to paired pulse stimulation of entorhinal and hippocampal areas. 2) coherence analysis of entorhinal/hippocampal seizure patterns. 3) analysis of frequency and periodicity of seizure onset. 4) measurement of early seizure propagation out of the limbic loop. Correlative analysis will be performed on a total of 100 surgical patients (50 retrospective, 50 prospective) with hippocampal, temporal neocortical, and entorhinal recordings of spontaneous seizures from implanted electrodes, analyzed by traditional statistical tests, and interpreted in light of independent measures of tissue pathophysiology, patient demographics, and surgical outcome.